Vehicle control device and vehicle control method

Abstract

A vehicle control device includes: a first behavior mode recognition unit that executes a first behavior mode recognition process of recognizing a first behavior mode of a target person based on an image captured by a first camera having a first resolution; a second behavior mode recognition unit that, when a distance between the vehicle and the target person is equal to or less than a predetermined distance, recognizes a second behavior mode of the target person based on an image captured by a second camera having a second resolution higher than the first resolution; a target person behavior recognition unit that recognizes a behavior of the target person against a vehicle based on the first behavior mode and the second behavior mode; and a control unit that executes predetermined control on the vehicle based on the behavior of the target person.

What is claimed is: 
     
       1. A vehicle control device comprising:
 a first front camera and a second front camera that are mounted on a vehicle and that capture a front of the vehicle, a first left side camera and a second left side camera that are mounted on the vehicle and that capture a left side of the vehicle, a first right side camera and a second right side camera that are mounted on the vehicle and that capture a right side of the vehicle, a first rear camera and a second rear camera that are mounted on the vehicle and that capture a rear of the vehicle, and a processor, 
 wherein 
 the second front camera, the second left side camera, the second right side camera, and the second rear camera correspond to the first front camera, the first left side camera, the first right side camera, and the first rear camera, respectively, and have resolutions higher than those of the first front camera, the first left side camera, the first right side camera, and the first rear camera, respectively, 
 the processor: 
 executes a first behavior mode recognition process of recognizing a movement locus of a target person existing around the vehicle from time-series captured images sequentially captured at a predetermined timing by the first front camera, the first left side camera, the first right side camera, and the first rear camera, and based on the movement locus recognized, recognizes a first behavior mode of the target person; 
 recognizes a distance between the vehicle and the target person; 
 when the distance between the vehicle and the target person is equal to or less than a predetermined distance during execution of the first behavior mode recognition process, executes a second behavior mode recognition process of starting operation of at least one of the second front camera, the second left side camera, the second right side camera, and the second rear camera which correspond to at least one of the first front camera, the first left side camera, the first right side camera, and the first rear camera that captures the captured images from which the target person is recognized, capturing surroundings of the vehicle by the at least one of the second front camera, the second left side camera, the second right side camera, and the second rear camera whose operation was started, and recognizing a second behavior mode of the target person based on an image captured by the at least one of the second front camera, the second left side camera, the second right side camera, and the second rear camera; 
 recognizes a behavior of the target person against the vehicle based on the first behavior mode and the second behavior mode; and 
 executes predetermined control on the vehicle based on the behavior of the target person recognized. 
 
     
     
       2. The vehicle control device according to  claim 1 , wherein
 the processor recognizes the first behavior mode having continuity with the second behavior mode in the first behavior mode recognition process by converting a position of an image portion of the target person in the image captured by the first camera into a corresponding position in the image captured by the second camera through coordinate conversion from a coordinate system of the first camera to a coordinate system of the second camera. 
 
     
     
       3. The vehicle control device according to  claim 1 , wherein
 the processor determines from a face image of the target person whether the target person is a user of the vehicle, the face image of the target person being extracted from the image captured by the second camera, and 
 regards the behavior of the target person recognized as a behavior of the user and executes the predetermined control when the target person is determined as the user. 
 
     
     
       4. The vehicle control device according to  claim 3 , wherein
 the processor sets a search range of an image of the target person in the image captured by the second camera, based on a position in the image captured by the first camera of an image portion of the target person extracted when the first behavior mode is recognized in the first behavior mode recognition process. 
 
     
     
       5. A vehicle control method to be executed by a computer, the method comprising:
 a first behavior mode recognition step of executing a first behavior mode recognition process of capturing surroundings of a vehicle by a first front camera and a second front camera that are mounted on a vehicle and that capture a front of the vehicle, a first left side camera and a second left side camera that are mounted on the vehicle and that capture a left side of the vehicle, a first right side camera and a second right side camera that are mounted on the vehicle and that capture a right side of the vehicle, a first rear camera and a second rear camera that are mounted on the vehicle and that capture a rear of the vehicle, the second front camera, the second left side camera, the second right side camera, and the second rear camera correspond to the first front camera, the first left side camera, the first right side camera, and the first rear camera, respectively, and have resolutions higher than those of the first front camera, the first left side camera, the first right side camera, and the first rear camera, respectively, recognizing a movement locus of a target person existing around the vehicle from time-series captured images sequentially captured at a predetermined timing by the first front camera, the first left side camera, the first right side camera, and the first rear camera, and based on the movement locus recognized, recognizing a first behavior mode of the target person; 
 a target person distance recognition step of recognizing a distance between the vehicle and the target person; 
 a second behavior mode recognition step of, when the distance between the vehicle and the target person is equal to or less than a predetermined distance during execution of the first behavior mode recognition process, executing a second behavior mode recognition process of starting operation of at least one of the second front camera, the second left side camera, the second right side camera, and the second rear camera, which correspond to at least one of the first front camera, the first left side camera, the first right side camera, and the first rear camera that captures the captured images from which the target person is recognized, capturing surroundings of the vehicle by the at least one of the second front camera, the second left side camera, the second right side camera, and the second rear camera whose operation was started, and recognizing a second behavior mode of the target person based on an image captured by the at least one of the second front camera, the second left side camera, the second right side camera, and the second rear camera; 
 a user behavior recognition step of recognizing a behavior of the target person against the vehicle based on the first behavior mode and the second behavior mode; and 
 a control step of executing predetermined control on the vehicle based on the behavior of the target person recognized in the user behavior recognition step.